Abstract
In recent years, there has been a clear paradigm shift in the field of production and logistics optimization, from batch processing to real-time stream processing of massive data volumes. The ability of a software system to adaptively adjust to changing conditions through integrated machine learning methods and heuristics is a tremendous advantage over competitors and can save costs or sustainably automate process optimization. Apache Kafka has established itself as the de facto standard for scalable, performant, and fault-tolerant processing of Big Data. Also in logistics optimization, optimization systems must be able to react as quickly as possible to changing conditions and carry out any necessary solution corrections. On the one hand, integrated machine learning methods are used again, on the other hand, the synchronization between the real world and the optimization model is of essential importance. This article explains two practical use cases and the architectures designed for them. Applying the described methods can solve the existing problems more efficiently and reliably.
Translated title of the contribution | Integrated Machine Learning for Real-Time Production Optimization and Decision-Making in Logistics |
---|---|
Original language | German (Austria) |
Title of host publication | Jahrbuch der Logistikforschung |
Subtitle of host publication | Innovative Anwendungen, Konzepte & Technologien |
Editors | Tina Wakolbinger |
Publisher | Trauner Verlag Linz |
Chapter | 3 |
Pages | 175-187 |
Number of pages | 12 |
Volume | 4 |
Edition | 1 |
ISBN (Electronic) | 978-3-99151-207-3 |
ISBN (Print) | 978-3-99151-207-3 |
Publication status | Published - 2023 |
Keywords
- Architecture
- Production Optimization
- Logistics Optimization
- Machine Learning
- Apache Kafka